Transgenic expression of CD95 ligand on thyroid follicular cells confers immune privilege upon thyroid allografts

Constitutive Fas ligand (FasL) expression by specialized cells in the body participates in the immune privilege status of tissues containing these cells. This property has been used to prevent rejection of allogeneic grafts. Nevertheless, the mechanism responsible for such protection has not been fully elucidated. Unfortunately, grafting of FasL transgenic (TG) tissues has been unsuccessful. We have generated TG mice expressing FasL (soluble + membrane bound) on thyroid follicular cells (TFC), and used them to show that ectopic FasL expression prevents thyroid allograft rejection. FasL expression on TFC led to markedly decreased anti-allogeneic, cytotoxic, and helper T lymphocyte activities. The alloantibody response in TG thyroid recipients was either completely inhibited or switched toward a T2-Ab response. Surprisingly, the beneficial effect of FasL on TG thyroid grafts was abolished by host CD4(+) T cell depletion. Host CD8(+) T cell depletion improved nontransgenic (NTG), but not TG graft survival. Altogether, our results suggest that FasL-induced tolerance is concomitant with a move away from a T1 type response, and a CD4 T cell-mediated regulation of the allocytotoxic T cell response. These results were dependent upon the level of FasL expression on TFC, in that low expression of FasL led to a less marked effect compared with the effect observed with high expression of FasL. These results provide some insight into the role of FasL in regulating destructive alloimmune responses in the case of whole organ grafting, and they have important implications for the development of FasL-based immunotherapy in organ transplantation.     

Cryptococcus neoformans capsular glucuronoxylomannan induces expression of fas ligand in macrophages

The major component of capsular material of Cryptococcus neoformans is glucuronoxylomannnan (GXM), a polysaccharide that exhibits potent immunosuppressive properties in vitro and in vivo. The results reported here show that 1) soluble purified GXM induces a prompt, long-lasting, and potent up-regulation of Fas ligand (FasL) on macrophages, 2) the up-regulation of FasL is related to induced synthesis and increased mobilization to the cellular surface, 3) this effect is largely mediated by interaction between GXM and TLR4, 4) FasL up-regulation occurs exclusively in GXM-loaded macrophages, 5) macrophages that show up-regulation of FasL induce apoptosis of activated T cells expressing Fas and Jurkat cells that constitutively express Fas, and 6) anti-Fas Abs rescue T cells from apoptosis induced by GXM. Collectively our results reveal novel aspects of the immunoregulatory properties of GXM and suggest that this nontoxic soluble compound could be used to dampen the immune response, to promote or accelerate the death receptor, and to fix FasL expression in a TLR/ligand-dependent manner. In the present study, we delineate potential new therapeutic applications for GXM that exploit death receptors as key molecular targets in regulating cell-mediated cytotoxicity, immune homeostasis, and the immunopathology of diseases.     

Induction of antitumor immunity with Fas/APO-1 ligand (CD95L)-transfected neuroblastoma neuro-2a cells.

Fas/Apo-1 (CD95)-Fas ligand (FasL) system has been implicated in the suppression and stimulation of immune responses. We examined the induction of antitumor immunity with neuroblastoma Neuro-2a cells transfected with FasL cDNA (Neuro-2a+FasL). Neuro-2a+FasL cells expressed FasL on the cell surface and secreted soluble FasL. Histologic and flow cytometric analyses revealed that Neuro-2a+FasL cells caused neutrophils to infiltrate into the injected site, resulting in strong inflammation. Neutrophil infiltration was inhibited by treatment with anti-FasL mAb and did not occur in Fas-deficient lpr mice. Normal syngeneic mice rejected Neuro-2a+FasL cells after the inflammation and acquired tumor-specific protective immunity. CD8+ T cells were responsible for the antitumor immunity. Neuro-2a+FasL cells formed tumors after far longer latency compared with mock-transfected Neuro-2a+Neo cells in nude mice, and immune competent mice rejected Neuro-2a cells but not sarcoma S713a cells when they were injected with Neuro-2a+FasL cells in a mixture. These results suggest that neutrophils attracted through the Fas-FasL system may impair tumor cells by inflammation at the initial step, followed by development of CD8+ T cell-dependent tumor-specific antitumor immunity, leading to complete eradication of tumor cells. Importantly, the treatment with Neuro-2a+FasL cells exhibited therapeutic efficacy against growing tumors.     

Membrane Fas ligand activates innate immunity and terminates ocular immune privilege

It has been proposed that the constitutive expression of Fas ligand (FasL) in the eye maintains immune privilege, in part through inducing apoptosis of infiltrating Fas(+) T cells. However, the role of FasL in immune privilege remains controversial due to studies that indicate FasL is both pro- and anti-inflammatory. To elucidate the mechanism(s) by which FasL regulates immune privilege, we used an ocular tumor model and examined the individual roles of the membrane-bound and soluble form of FasL in regulating ocular inflammation. Following injection into the privileged eye, tumors expressing only soluble FasL failed to trigger inflammation and grew progressively. By contrast, tumors expressing only membrane FasL 1) initiated vigorous neutrophil-mediated inflammation, 2) terminated immune privilege, and 3) were completely rejected. Moreover, the rejection coincided with activation of both innate and adaptive immunity. Interestingly, a higher threshold level of membrane FasL on tumors is required to initiate inflammation within the immune privileged eye, as compared with nonprivileged sites. The higher threshold is due to the suppressive microenvironment found within aqueous humor that blocks membrane FasL activation of neutrophils. However, aqueous humor is unable to completely block the proinflammatory effects of tumor cells that express high levels of membrane FasL. In conclusion, our data indicate that the function of FasL on intraocular tumors is determined by the microenvironment in conjunction with the form and level of FasL expressed.     

Mature but not immature Fas ligand (CD95L)-transduced human monocyte-derived dendritic cells are protected from Fas-mediated apoptosis and can be used as killer APC

Several in vitro and animal studies have been performed to modulate the interaction of APCs and T cells by Fas (CD95/Apo-1) signaling to delete activated T cells in an Ag-specific manner. However, due to the difficulties in vector generation and low transduction frequencies, similar studies with primary human APC are still lacking. To evaluate whether Fas ligand (FasL/CD95L) expressing killer APC could be generated from primary human APC, monocyte-derived dendritic cells (DC) were transduced using the inducible Cre/Loxp adenovirus vector system. Combined transduction of DC by AdLoxpFasL and AxCANCre, but not single transduction with these vectors, resulted in dose- and time-dependent expression of FasL in >70% of mature DC (mDC), whereas <20% of immature DC (iDC) expressed FasL. In addition, transduction by AdLoxpFasL and AxCANCre induced apoptosis in >80% of iDC, whereas FasL-expressing mDC were protected from FasL/Fas (CD95/Apo-1)-mediated apoptosis despite coexpression of Fas. FasL-expressing mDC eliminated Fas(+) Jurkat T cells as well as activated primary T cells by apoptosis, whereas nonactivated primary T cells were not deleted. Induction of apoptosis in Fas(+) target cells required expression of FasL in DC and cell-to-cell contact between effector and target cell, and was not dependent on soluble FasL. Induction of apoptosis in Fas(+) target cells required expression of FasL in DC, cell-to-cell contact between effector and target cell, and was not dependent on soluble FasL. The present results demonstrate that FasL-expressing killer APC can be generated from human monocyte-derived mDC using adenoviral gene transfer. Our results support the strategy to use killer APCs as immunomodulatory cells for the treatment of autoimmune disease and allograft rejection.     

An essential contribution by IFN-gamma to CD8+ T cell-mediated rejection of pancreatic islet allografts

CD8+ T cells have long been considered to be the prototypical cytotoxic lymphocyte subpopulation. However, whether alloreactive CD8+ T cells require traditional cytolytic pathways such as perforin and Fas ligand (FasL) to mediate graft rejection has been a controversial issue. In the present studies, we examined the role of varied effector pathways in CD8+ T cell-mediated rejection of pancreatic islet allografts. Our goal was to systematically determine the relative requirements, if any, of perforin and FasL as well as the proinflammatory cytokine IFN-gamma in triggering graft destruction. To study CD8+ T cell effector pathways independently of other lymphocyte populations, purified alloreactive CD8+ T cells were adoptively transferred into severe combined immune-deficient (SCID) recipients bearing established islet allografts. Results indicate that to reject established islet allografts, primed CD8+ T cells do not require the individual action of the conventional cytotoxic effectors perforin and Fas ligand. In contrast, the ability to produce IFN-gamma is critical for efficient CD8+ T cell-mediated rejection of established islet allografts. Furthermore, alloreactive CD8+ TCR transgenic T cells (2C) also show IFN-gamma dependence for mediating islet allograft rejection in vivo. We speculate from these results that the production of IFN-gamma by alloreactive CD8+ T cells is a rate-limiting step in the process of islet allograft rejection.     

Processing of tumor necrosis factor by the membrane-bound TNF-alpha-converting enzyme, but not its truncated soluble form.

Tumour necrosis factor (TNF)-alpha-converting enzyme (TACE) is a membrane protein belonging to the ADAM (a disintegrin and metalloproteinase) family that cleaves various membrane proteins, including the proform of TNF-alpha. In this study, we constructed expression vectors for the membrane-bound full-length TACE (mTACE) and its truncated soluble form (sTACE). When a human TNF-alpha expression vector was introduced into human 293 cells, processing of TNF-alpha to its mature form was enhanced by coexpressing mTACE, and this processing was inhibited by a metalloproteinase inhibitor. On the other hand, coexpression of sTACE had no effect on the processing of TNF-alpha, although the culture medium of sTACE-transfected cells could cleave a peptide containing the TNF-alpha cleavage site. Fas ligand (FasL)-transfected 293 cells released a considerable amount of soluble FasL, and coexpression of neither mTACE nor sTACE enhanced this shedding. Immunoprecipitation and Western blotting analysis with cells that were cotransfected with TACE and TNF-alpha indicated that both mTACE and sTACE could interact with the proform of TNF-alpha. In the same assay, neither mTACE nor sTACE interacted with FasL. The catalytic domain-lacking TACE mutant, which could also interact TNF-alpha, showed a dominant negative effect on not only TNF-alpha secretion but also FasL secretion. These results suggest that binding of the membrane-anchored but not the soluble form of TACE to TNF-alpha results in efficient ectodomain shedding, and that FasL secretase is a metalloproteinase similar, but not identical, to TACE.     

Presence of Fas-Fas ligand system and bcl-2 gene products in cells and fluids from gonadotropin-stimulated human ovaries

Apoptosis, or programmed cell death, is an important mechanism for the regulation of embryonic development and tissue homeostasis. It is coordinated by a number of molecules including the Fas-Fas ligand (FasL) system and bcl-2. The purpose of this study was to characterize the expression of these molecules in human oocytes and cumulus cells from gonadotropin-stimulated human ovaries and to determine whether the presence of soluble Fas (sFas), soluble FasL, or interferon-gamma in follicular fluid (FF) correlated with apoptosis in cumulus cells, oocyte maturation, and embryo quality. Levels of sFas were significantly higher in FF containing immature oocytes compared with those containing atretic oocytes (P < 0.05; FF containing mature oocytes had highly variable levels of sFas. Levels of sFas in FF did not correlate with either fertilization, embryo quality resulting from fertilized oocytes, or apoptosis rate in cumulus cells. Fas was expressed in both unfertilized oocytes and cumulus cells, whereas FasL expression was not usually detected in these cell types. Messenger RNA for bcl-2 was detectable in both freshly isolated oocytes and cumulus cells but was not demonstrable following 24 h of culture that coincided with a significant increase of apoptosis in cumulus cells. Our results indicate that soluble forms of the Fas-FasL system are present in FF from gonadotropin-stimulated human ovaries and suggest that this system may play a role in preventing oocyte atresia during folliculogenesis but is probably not important for apoptotic events in cumulus cells and oocytes after fertilization failure. Apoptosis in this case may be facilitated by the downregulation of bcl-2. Further studies on the expression of these molecules in follicles containing atretic oocytes and immature oocytes are needed to confirm this new hypothesis.     

Palmitoylation of human FasL modulates its cell death-inducing function

Fas ligand (FasL) is a transmembrane protein that regulates cell death in Fas-bearing cells. FasL-mediated cell death is essential for immune system homeostasis and the elimination of viral or transformed cells. Because of its potent cytotoxic activity, FasL expression at the cell surface is tightly regulated, for example, via processing by ADAM10 and SPPL2a generating soluble FasL and the intracellular fragments APL (ADAM10-processed FasL form) and SPA (SPPL2a-processed APL). In this study, we report that FasL processing by ADAM10 counteracts Fas-mediated cell death and is strictly regulated by membrane localization, interactions and modifications of FasL. According to our observations, FasL processing occurs preferentially within cholesterol and sphingolipid-rich nanodomains (rafts) where efficient Fas–FasL contact occurs, Fas receptor and FasL interaction is also required for efficient FasL processing, and FasL palmitoylation, which occurs within its transmembrane domain, is critical for efficient FasL-mediated killing and FasL processing.     

Type I transforming growth factor-beta receptors on neutrophils mediate chemotaxis to transforming growth factor-beta.

Participation of human polymorphonuclear neutrophils in the inflammatory response is mediated, in part, by soluble factors such as chemotactic peptides and cytokines. Although the cytokine, transforming growth factor beta (TGF-beta), has been shown to recruit monocytes and promote the inflammatory process, its effects on neutrophils are unknown. In this investigation, [125I]TGF-beta 1 affinity binding studies were employed to show that neutrophils express TGF-beta receptors (350 +/- 20 receptors/cell), which exhibit high affinity for the ligand (dissociation constant, 50 pM). Affinity cross-linking studies identified the receptors to be primarily of the type I class. In contrast to the receptors on monocytes, neutrophil TGF-beta receptors were not down-regulated by exposure to specific inflammatory mediators. Additional studies examined whether exposure of neutrophils to TGF-beta could enhance specific functions, as occurs with monocytes. TGF-beta was shown to cause directed migration of neutrophils at femtomolar concentrations, thus it is the most potent neutrophil chemotactic factor yet identified. Neutrophil production of reactive oxygen intermediates was not stimulated by TGF-beta, nor did TGF-beta enhance or depress subsequent PMA- or FMLP-stimulated superoxide production. However, the stable expression of neutrophil TGF-beta receptors, and the capacity of this cytokine to stimulate neutrophil chemotaxis, suggest that the pro-inflammatory effects of TGF-beta are mediated by neutrophils in addition to monocytes.     

Bioactivities of Fas ligand-expressing retroviral particles

Culture supernatants from retroviral packaging cells carrying the human Fas ligand (FasL) gene killed both human (Jurkat) and mouse (LB27.4) targets within 5 h of incubation. Cytotoxicity was found both in a fraction >/=500 kDa and a fraction between 50 and 500 kDa. Following ultracentrifugation, the activity in the >/=500-kDa fraction was concentrated in the pellet (FasL vector preparation (VP)), which was also infective when added to NIH-3T3 cells. Both Polybrene and poly-l -lysine significantly enhanced the cytotoxicity of FasL VP but not anti-Fas mAb, soluble FasL (sFasL), and cell-associated FasL. In the presence of Polybrene, FasL VP killed targets that are resistant to anti-Fas mAb and sFasL. The infectivity but not FasL cytotoxicity of FasL VP was sensitive to irradiation and heat shock. By contrast, cytotoxicity of FasL VP could be enhanced or inhibited depending on the doses of anti-FasL mAb. Interestingly, the infectivity of FasL VP was specifically enhanced by anti-FasL mAb, suggesting that a nonviral gene product could be used to regulate the behavior of the retroviral vector. Thus, in addition to expressing potent FasL cytotoxicity, the FasL VP exhibits unique properties heretofore not attributed to anti-Fas mAb, sFasL, and cell-associated FasL. Our study raises the possibility of using the retroviral gene-packaging technology to make powerful, versatile, and regulatable bioactive vesicles expressing a predetermined function of the protein encoded by the target gene.     

Depletion of resident alveolar macrophages does not prevent Fas-mediated lung injury in mice

Activation of the Fas/Fas ligand (FasL) system in the lungs results in a form of injury characterized by alveolar epithelial apoptosis and neutrophilic inflammation. Studies in vitro show that Fas activation induces apoptosis in alveolar epithelial cells and cytokine production in alveolar macrophages. The main goal of this study was to determine the contribution of alveolar macrophages to Fas-induced lung inflammation in mice, by depleting alveolar macrophages using clodronate-containing liposomes. Liposomes containing clodronate or PBS were instilled by intratracheal instillation. After 24 h, the mice received intratracheal instillations of the Fas-activating monoclonal antibody Jo2 or an isotype control antibody and were studied 18 h later. The Jo2 MAb induced increases in bronchoalveolar lavage fluid (BALF) total neutrophils, lung caspase-3 activity, and BALF total protein and worsened histological lung injury in the macrophage-depleted mice. Studies in vitro showed that Fas activation induced the release of the cytokine KC in a mouse lung epithelial cell line, MLE-12. These results suggest that the lung inflammatory response to Fas activation is not primarily dependent on resident alveolar macrophages and may instead depend on cytokine release by alveolar epithelial cells.     

Functional Characterization of a Chimeric Soluble Fas Ligand Polymer with In Vivo Anti-Tumor Activity

Binding of ligand FasL to its receptor Fas triggers apoptosis via the caspase cascade. FasL itself is homotrimeric, and a productive apoptotic signal requires that FasL be oligomerized beyond the homotrimeric state. We generated a series of FasL chimeras by fusing FasL to domains of the Leukemia Inhibitory Factor receptor gp190 which confer homotypic oligomerization, and analyzed the capacity of these soluble chimeras to trigger cell death. We observed that the most efficient FasL chimera, called pFasL, was also the most polymeric, as it reached the size of a dodecamer. Using a cellular model, we investigated the structure-function relationships of the FasL/Fas interactions for our chimeras, and we demonstrated that the Fas-mediated apoptotic signal did not solely rely on ligand-mediated receptor aggregation, but also required a conformational adaptation of the Fas receptor. When injected into mice, pFasL did not trigger liver injury at a dose which displayed anti-tumor activity in a model of human tumor transplanted to immunodeficient animals, suggesting a potential therapeutic use. Therefore, the optimization of the FasL conformation has to be considered for the development of efficient FasL-derived anti-cancer drugs targeting Fas.     

Increased cytotoxicity of soluble Fas ligand by fusing isoleucine zipper motif

Fas (CD95) ligand (FasL) has the ability to induce apoptosis in Fas-expressing glioma cells by binding to Fas. Several molecular species have been designed to be soluble Fas ligands for therapeutic purposes. We successfully constructed a chimeric soluble FasL by fusing an isoleucine zipper motif for self-oligomerization and a FLAG sequence to the extracellular domain of the human Fas ligand (FIZ-shFasL). The cytotoxic effect of FIZ-shFasL on Jurkat cells was equivalent to that of membrane-bound FasL and approximately 10-fold stronger than that of agonistic anti-Fas antibody (CH-11). Flow cytometric analysis demonstrated that the differential Fas expression of human brain tumor cell lines partially correlated with levels of apoptosis through FIZ-shFasL. The upper limit of FIZ-shFasL for safe systemic administration to rat is estimated as below 2 microg/ml in plasma concentration. FIZ-shFasL could be applicable as a therapeutic agent for cancer.     

Neutrophil expression of fas ligand and perforin directs effector CD8 T cell infiltration into antigen-challenged skin

Contact hypersensitivity (CHS) is a T cell response to hapten skin challenge of sensitized individuals proposed to be mediated by hapten-primed CD8 cytolytic T cells. Effector CD8 T cell recruitment into hapten challenge sites to elicit CHS requires prior CXCL1- and CXCL2-mediated neutrophil infiltration into the site. We investigated whether neutrophil activities directing hapten-primed CD8 T cell skin infiltration in response to 2,4-dinitro-1-fluorobenzene (DNFB) required Fas ligand (FasL) and perforin expression. Although DNFB sensitization of gld/perforin(-/-) mice induced hapten-specific CD8 T cells producing IFN-γ and IL-17, these T cells did not infiltrate the DNFB challenge site to elicit CHS but did infiltrate the challenge site and elicit CHS when transferred to hapten-challenged naive wild-type recipients. Hapten-primed wild-type CD8 T cells, however, did not elicit CHS when transferred to naive gld/perforin(-/-) recipients. Wild-type bone marrow neutrophils expressed FasL and perforin, and when transferred to sensitized gld/perforin(-/-) mice, they restored hapten-primed CD8 T cell infiltration into the challenge site and CHS. The FasL/perforin-mediated activity of wild-type neutrophils induced the expression of T cell chemoattractants, CCL1, CCL2, and CCL5, within the hapten-challenged skin. These results indicate FasL/perforin-independent functions of hapten-primed CD8 T cells in CHS and identify new functions for neutrophils in regulating effector CD8 T cell recruitment and immune responses in the skin.     

Cutting edge: the tumor counterattack hypothesis revisited: colon cancer cells do not induce T cell apoptosis via the Fas (CD95, APO-1) pathway

The counterattack hypothesis, suggesting that cancer cells express Fas ligand (FasL) and are able to kill Fas-expressing tumor-infiltrating activated T cells, was supported by reports of the killing of Jurkat cells by FasL-expressing human colon cancer cell lines. Through the use of an improved cytotoxic assay in which soluble FasL and FasL-transfected KFL9 cells were used as positive controls, we show that none of seven human colon cancer cell lines induce apoptosis of two Fas-expressing target cell lines, Jurkat and L1210-Fas cells. Moreover, in coculture experiments, cancer cell monolayers do not inhibit the growth of Fas-expressing lymphoid cells. Although FasL mRNA and protein were detected in the extracts of the colon cancer cell lines, flow cytometry and confocal microscopy failed to detect the protein on the surface of tumor cells. These results suggest that the counterattack of tumor-infiltrating T lymphocytes by cancer cells may not account for immune tolerance toward tumor cells.     

ANG II induces apoptosis of human vascular smooth muscle via extrinsic pathway involving inhibition of Akt phosphorylation and increased FasL expression

In addition to well-documented vascular growth-promoting effects, ANG II exerts proapoptotic effects that are poorly understood. IGF-1 is a potent survival factor for human vascular smooth muscle cells (hVSMC), and its antiapoptotic effects are mediated via the IGF-1 receptor (IGF-1R) through a signaling pathway involving phosphatidylinositol 3-kinase and Akt. We hypothesized that there would be cross talk between ANG II proapoptotic effects and IGF-1 survival effects in hVSMC. To investigate ANG II-induced apoptosis and the potential involvement of IGF-1, we exposed quiescent and nonquiescent hVSMC to ANG II. ANG II induced apoptosis only in nonquiescent cells but stimulated hypertrophy in quiescent cells. ANG II-induced apoptosis was characterized by marked inhibition of Akt phosphorylation and stimulation of membrane Fas ligand (FasL) expression, caspase-8 activation, and a reduction in soluble FasL expression. Adenovirally mediated overexpression of Akt rescued hVSMC from ANG II-induced apoptosis. IGF-1R activation increased Akt phosphorylation and soluble FasL expression, and these effects were completely blocked by coincubating hVSMC with ANG II. In conclusion, ANG II-induced apoptosis of hVSMC is characterized by marked inhibition of Akt phosphorylation and stimulation of an extrinsic cell death signaling pathway via upregulation of membrane FasL expression, caspase-8 activation, and a reduction in soluble FasL expression. Furthermore, ANG II antagonizes the antiapoptotic effect of IGF-1 by blocking its ability to increase Akt phosphorylation and soluble FasL. These findings provide novel insights into ANG II-induced apoptotic signaling and have significant implication for understanding ANG II-induced remodeling in hypertension and atherosclerosis.     

Expression of the functional soluble form of human fas ligand in activated lymphocytes.

Fas is a type I membrane protein which mediates apoptosis. Fas ligand (FasL) is a 40 kDa type II membrane protein expressed in cytotoxic T cells upon activation that belongs to the tumor necrosis factor (TNF) family. Here, we found abundant cytotoxic activity against Fas-expressing cells in the supernatant of COS cells transfected with human FasL cDNA but not with murine FasL cDNA. Using a specific polyclonal antibody against a peptide in the extracellular region of human FasL, a protein of 26 kDa was detected in the supernatant of the COS cells. The signal sequence of granulocyte colony-stimulating factor was attached to the extracellular region of human FasL. COS cells transfected with the cDNA coding for the chimeric protein efficiently secreted the active soluble form of human FasL (sFasL). Chemical crosslinking and gel filtration analysis suggested that human sFasL exists as a trimer. Human peripheral T cells activated with phorbol myristic acetate and ionomycin also produced functional sFasL, suggesting that human sFasL works as a pathological agent in systemic tissue injury.     

Possible role of organ-specific autoantigen for Fas ligand-mediated activation-induced cell death in murine Sj?gren's syndrome.

Activation-induced cell death (AICD) is a well-known mechanism of peripheral T cell tolerance that depends upon an interaction between Fas and Fas ligand (FasL). In this study, we demonstrate that the administration of a soluble form of anti-FasL Ab, FLIM58, results in severe destructive autoimmune exocrinopathy in the murine model of human Sj?gren's syndrome (SS), and we found that an organ-specific autoantigen may play an important role on down-modulation of AICD. A high titer of serum autoantibodies against 120-kDa alpha-fodrin autoantigen was detected in the FLIM58-treated mice, and splenic T cell culture supernatants contained high levels of IFN-gamma. In vitro T cell apoptosis assay indicated that FasL-mediated AICD is down-regulated by autoantigen stimulation in spleen cells from the murine SS model, but not from Fas-deficient MRL/lpr mice and FasL-deficient MRL/gld mice. FasL undergo metalloproteinase-mediated proteolytic processing in their extracellular domains, resulting in the release of soluble trimeric ligands (soluble FasL). We showed that the processing of soluble FasL occurs in autoantigen-specific CD4(+) T cells, and that a significant increase in expressions of metalloproteinase-9 mRNA was observed in spleen cells from SS model mice. These findings indicate that the increased generation of soluble FasL inhibits the normal AICD process, leading to the proliferation of effector CD4(+) T cells in the murine SS model.